/dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/arm/ |
H A D | convolutiondepthwise_5x5_pack4.h | 232 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 260 _sum00 = vmlaq_f32(_sum00, _k32, _r32); in convdw5x5s1_pack4_neon() 265 _sum01 = vmlaq_f32(_sum01, _k32, _r33); in convdw5x5s1_pack4_neon() 498 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 675 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 688 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 707 _sum1 = vmlaq_f32(_sum1, _k32, _r42); in convdw5x5s1_pack4_neon() 886 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 893 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 898 _sum1 = vmlaq_f32(_sum1, _k32, _r33); in convdw5x5s1_pack4_neon() [all …]
|
H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1378 float16x8_t _k32 = vld1q_f16(k0 + 16); in convdw5x5s2_pack8_fp16sa_neon() local 1385 _sum0 = vfmaq_f16(_sum0, _k32, _r32); in convdw5x5s2_pack8_fp16sa_neon()
|
/dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/arm/ |
H A D | convolutiondepthwise_5x5_pack4.h | 232 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 260 _sum00 = vmlaq_f32(_sum00, _k32, _r32); in convdw5x5s1_pack4_neon() 265 _sum01 = vmlaq_f32(_sum01, _k32, _r33); in convdw5x5s1_pack4_neon() 498 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 675 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 688 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 707 _sum1 = vmlaq_f32(_sum1, _k32, _r42); in convdw5x5s1_pack4_neon() 886 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 893 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 898 _sum1 = vmlaq_f32(_sum1, _k32, _r33); in convdw5x5s1_pack4_neon() [all …]
|
H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1378 float16x8_t _k32 = vld1q_f16(k0 + 16); in convdw5x5s2_pack8_fp16sa_neon() local 1385 _sum0 = vfmaq_f16(_sum0, _k32, _r32); in convdw5x5s2_pack8_fp16sa_neon()
|
/dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/arm/ |
H A D | convolutiondepthwise_5x5_pack4.h | 232 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 260 _sum00 = vmlaq_f32(_sum00, _k32, _r32); in convdw5x5s1_pack4_neon() 265 _sum01 = vmlaq_f32(_sum01, _k32, _r33); in convdw5x5s1_pack4_neon() 498 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 675 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 688 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 707 _sum1 = vmlaq_f32(_sum1, _k32, _r42); in convdw5x5s1_pack4_neon() 886 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 893 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 898 _sum1 = vmlaq_f32(_sum1, _k32, _r33); in convdw5x5s1_pack4_neon() [all …]
|
H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1378 float16x8_t _k32 = vld1q_f16(k0 + 16); in convdw5x5s2_pack8_fp16sa_neon() local 1385 _sum0 = vfmaq_f16(_sum0, _k32, _r32); in convdw5x5s2_pack8_fp16sa_neon()
|
/dports/misc/ncnn/ncnn-20211208/src/layer/arm/ |
H A D | convolutiondepthwise_5x5_pack4.h | 232 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 260 _sum00 = vmlaq_f32(_sum00, _k32, _r32); in convdw5x5s1_pack4_neon() 265 _sum01 = vmlaq_f32(_sum01, _k32, _r33); in convdw5x5s1_pack4_neon() 498 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 675 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 688 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 707 _sum1 = vmlaq_f32(_sum1, _k32, _r42); in convdw5x5s1_pack4_neon() 886 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 893 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 898 _sum1 = vmlaq_f32(_sum1, _k32, _r33); in convdw5x5s1_pack4_neon() [all …]
|
H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1378 float16x8_t _k32 = vld1q_f16(k0 + 16); in convdw5x5s2_pack8_fp16sa_neon() local 1385 _sum0 = vfmaq_f16(_sum0, _k32, _r32); in convdw5x5s2_pack8_fp16sa_neon()
|
/dports/graphics/realsr-ncnn-vulkan/realsr-ncnn-vulkan-20210210/src/ncnn/src/layer/arm/ |
H A D | convolutiondepthwise_5x5_pack4.h | 232 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 260 _sum00 = vmlaq_f32(_sum00, _k32, _r32); in convdw5x5s1_pack4_neon() 265 _sum01 = vmlaq_f32(_sum01, _k32, _r33); in convdw5x5s1_pack4_neon() 498 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 675 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 688 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 707 _sum1 = vmlaq_f32(_sum1, _k32, _r42); in convdw5x5s1_pack4_neon() 886 float32x4_t _k32 = vld1q_f32(k0 + 8); in convdw5x5s1_pack4_neon() local 893 _sum0 = vmlaq_f32(_sum0, _k32, _r32); in convdw5x5s1_pack4_neon() 898 _sum1 = vmlaq_f32(_sum1, _k32, _r33); in convdw5x5s1_pack4_neon() [all …]
|
H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1378 float16x8_t _k32 = vld1q_f16(k0 + 16); in convdw5x5s2_pack8_fp16sa_neon() local 1385 _sum0 = vfmaq_f16(_sum0, _k32, _r32); in convdw5x5s2_pack8_fp16sa_neon()
|
/dports/misc/ncnn/ncnn-20211208/src/layer/mips/ |
H A D | convolutiondepthwise_5x5_pack4.h | 148 v4f32 _k32 = (v4f32)__msa_ld_w(k0 + 4 * 2, 0); in convdw5x5s1_pack4_msa() local 155 _sum0 = __msa_fmadd_w(_sum0, _k32, _r32); in convdw5x5s1_pack4_msa() 167 _sum1 = __msa_fmadd_w(_sum1, _k32, _r42); in convdw5x5s1_pack4_msa() 300 v4f32 _k32 = (v4f32)__msa_ld_w(k0 + 4 * 2, 0); in convdw5x5s1_pack4_msa() local 307 _sum0 = __msa_fmadd_w(_sum0, _k32, _r32); in convdw5x5s1_pack4_msa() 463 v4f32 _k32 = (v4f32)__msa_ld_w(k0 + 4 * 2, 0); in convdw5x5s2_pack4_msa() local 470 _sum0 = __msa_fmadd_w(_sum0, _k32, _r32); in convdw5x5s2_pack4_msa()
|
H A D | convolution_7x7_pack1to4.h | 237 v4f32 _k32 = (v4f32)__msa_ld_w(kptr + 4 * 2, 0); in conv7x7s2_pack1to4_msa() local 271 _sum0 = __msa_fmadd_w(_sum0, _r32, _k32); in conv7x7s2_pack1to4_msa() 272 _sum1 = __msa_fmadd_w(_sum1, _r34, _k32); in conv7x7s2_pack1to4_msa() 273 _sum2 = __msa_fmadd_w(_sum2, _r36, _k32); in conv7x7s2_pack1to4_msa() 274 _sum3 = __msa_fmadd_w(_sum3, _r38, _k32); in conv7x7s2_pack1to4_msa() 547 v4f32 _k32 = (v4f32)__msa_ld_w(kptr + 4 * 2, 0); in conv7x7s2_pack1to4_msa() local 560 _sum0 = __msa_fmadd_w(_sum0, (v4f32)__msa_splati_w(_r3, 2), _k32); in conv7x7s2_pack1to4_msa()
|
/dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/x86/ |
H A D | convolutiondepthwise_5x5_pack8.h | 116 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s1_pack8_avx() local 123 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s1_pack8_avx() 270 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s2_pack8_avx() local 277 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s2_pack8_avx()
|
/dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/x86/ |
H A D | convolutiondepthwise_5x5_pack8.h | 116 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s1_pack8_avx() local 123 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s1_pack8_avx() 270 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s2_pack8_avx() local 277 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s2_pack8_avx()
|
/dports/misc/ncnn/ncnn-20211208/src/layer/x86/ |
H A D | convolutiondepthwise_5x5_pack8.h | 116 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s1_pack8_avx() local 123 _sum0 = _mm256_comp_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s1_pack8_avx() 270 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s2_pack8_avx() local 277 _sum0 = _mm256_comp_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s2_pack8_avx()
|
/dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/x86/ |
H A D | convolutiondepthwise_5x5_pack8.h | 116 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s1_pack8_avx() local 123 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s1_pack8_avx() 270 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s2_pack8_avx() local 277 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s2_pack8_avx()
|
/dports/graphics/realsr-ncnn-vulkan/realsr-ncnn-vulkan-20210210/src/ncnn/src/layer/x86/ |
H A D | convolutiondepthwise_5x5_pack8.h | 116 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s1_pack8_avx() local 123 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s1_pack8_avx() 270 __m256 _k32 = _mm256_loadu_ps(k0 + 16); in convdw5x5s2_pack8_avx() local 277 _sum0 = _mm256_fmadd_ps(_k32, _r32, _sum0); in convdw5x5s2_pack8_avx()
|
/dports/misc/ncnn/ncnn-20211208/src/layer/riscv/ |
H A D | convolutiondepthwise_5x5_packn.h | 142 vfloat32m1_t _k32 = vle32_v_f32m1(k0 + packn * 2, vl); in convdw5x5s1_packn_rvv() local 149 _sum0 = vfmacc_vv_f32m1(_sum0, _k32, _r32, vl); in convdw5x5s1_packn_rvv() 161 _sum1 = vfmacc_vv_f32m1(_sum1, _k32, _r42, vl); in convdw5x5s1_packn_rvv() 286 vfloat32m1_t _k32 = vle32_v_f32m1(k0 + packn * 2, vl); in convdw5x5s1_packn_rvv() local 293 _sum0 = vfmacc_vv_f32m1(_sum0, _k32, _r32, vl); in convdw5x5s1_packn_rvv() 444 vfloat32m1_t _k32 = vle32_v_f32m1(k0 + packn * 2, vl); in convdw5x5s2_packn_rvv() local 451 _sum0 = vfmacc_vv_f32m1(_sum0, _k32, _r32, vl); in convdw5x5s2_packn_rvv()
|
H A D | convolutiondepthwise_5x5_packn_fp16s.h | 142 vfloat16m1_t _k32 = vle16_v_f16m1(k0 + packn * 2, vl); in convdw5x5s1_packn_fp16sa_rvv() local 149 _sum0 = vfmacc_vv_f16m1(_sum0, _k32, _r32, vl); in convdw5x5s1_packn_fp16sa_rvv() 161 _sum1 = vfmacc_vv_f16m1(_sum1, _k32, _r42, vl); in convdw5x5s1_packn_fp16sa_rvv() 286 vfloat16m1_t _k32 = vle16_v_f16m1(k0 + packn * 2, vl); in convdw5x5s1_packn_fp16sa_rvv() local 293 _sum0 = vfmacc_vv_f16m1(_sum0, _k32, _r32, vl); in convdw5x5s1_packn_fp16sa_rvv() 444 vfloat16m1_t _k32 = vle16_v_f16m1(k0 + packn * 2, vl); in convdw5x5s2_packn_fp16sa_rvv() local 451 _sum0 = vfmacc_vv_f16m1(_sum0, _k32, _r32, vl); in convdw5x5s2_packn_fp16sa_rvv()
|
H A D | convolution_7x7_pack1ton.h | 298 _sum0 = vfmacc_vf_f32m1(_sum0, r3[2], _k32, vl); in conv7x7s2_pack1ton_rvv() 299 _sum1 = vfmacc_vf_f32m1(_sum1, r3[4], _k32, vl); in conv7x7s2_pack1ton_rvv() 300 _sum2 = vfmacc_vf_f32m1(_sum2, r3[6], _k32, vl); in conv7x7s2_pack1ton_rvv() 301 _sum3 = vfmacc_vf_f32m1(_sum3, r3[8], _k32, vl); in conv7x7s2_pack1ton_rvv() 302 _sum4 = vfmacc_vf_f32m1(_sum4, r3[10], _k32, vl); in conv7x7s2_pack1ton_rvv() 303 _sum5 = vfmacc_vf_f32m1(_sum5, r3[12], _k32, vl); in conv7x7s2_pack1ton_rvv() 304 _sum6 = vfmacc_vf_f32m1(_sum6, r3[14], _k32, vl); in conv7x7s2_pack1ton_rvv() 305 _sum7 = vfmacc_vf_f32m1(_sum7, r3[16], _k32, vl); in conv7x7s2_pack1ton_rvv() 701 _sum0 = vfmacc_vf_f32m1(_sum0, r3[2], _k32, vl); in conv7x7s2_pack1ton_rvv() 702 _sum1 = vfmacc_vf_f32m1(_sum1, r3[4], _k32, vl); in conv7x7s2_pack1ton_rvv() [all …]
|
H A D | convolution_7x7_pack1ton_fp16s.h | 298 _sum0 = vfmacc_vf_f16m1(_sum0, r3[2], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 299 _sum1 = vfmacc_vf_f16m1(_sum1, r3[4], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 300 _sum2 = vfmacc_vf_f16m1(_sum2, r3[6], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 301 _sum3 = vfmacc_vf_f16m1(_sum3, r3[8], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 302 _sum4 = vfmacc_vf_f16m1(_sum4, r3[10], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 303 _sum5 = vfmacc_vf_f16m1(_sum5, r3[12], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 304 _sum6 = vfmacc_vf_f16m1(_sum6, r3[14], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 305 _sum7 = vfmacc_vf_f16m1(_sum7, r3[16], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 701 _sum0 = vfmacc_vf_f16m1(_sum0, r3[2], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() 702 _sum1 = vfmacc_vf_f16m1(_sum1, r3[4], _k32, vl); in conv7x7s2_pack1ton_fp16sa_rvv() [all …]
|
/dports/lang/python-legacy/Python-2.7.18/Lib/test/support/ |
H A D | __init__.py | 1987 self._k32 = ctypes.windll.kernel32 1989 self.old_value = self._k32.SetErrorMode(SEM_NOGPFAULTERRORBOX) 1990 self._k32.SetErrorMode(self.old_value | SEM_NOGPFAULTERRORBOX) 2052 self._k32.SetErrorMode(self.old_value)
|
/dports/lang/python27/Python-2.7.18/Lib/test/support/ |
H A D | __init__.py | 1987 self._k32 = ctypes.windll.kernel32 1989 self.old_value = self._k32.SetErrorMode(SEM_NOGPFAULTERRORBOX) 1990 self._k32.SetErrorMode(self.old_value | SEM_NOGPFAULTERRORBOX) 2052 self._k32.SetErrorMode(self.old_value)
|
/dports/lang/python37/Python-3.7.12/Lib/test/support/ |
H A D | __init__.py | 2630 self._k32 = ctypes.windll.kernel32 2632 self.old_value = self._k32.SetErrorMode(SEM_NOGPFAULTERRORBOX) 2633 self._k32.SetErrorMode(self.old_value | SEM_NOGPFAULTERRORBOX) 2689 self._k32.SetErrorMode(self.old_value)
|
/dports/devel/py-pyzipper/pyzipper-0.3.5/test/support/ |
H A D | __init__.py | 2615 self._k32 = ctypes.windll.kernel32 2617 self.old_value = self._k32.SetErrorMode(SEM_NOGPFAULTERRORBOX) 2618 self._k32.SetErrorMode(self.old_value | SEM_NOGPFAULTERRORBOX) 2674 self._k32.SetErrorMode(self.old_value)
|